AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Intra-Modal Proxy Learning for Zero-Shot Visual Categorization with CLIP

Qi Qian, Yuanhong Xu, Juhua Hu

NeurIPS 2023 · Vision-language and multimodal › CLIP and zero-shot · approach: Attribute-based zero-shot learning · 4 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
learn intra-modal vision proxy (InMaP) instead of text proxyCLIP zero-shot · ImageNet ResNet-50accuracy (%): 60.32 -> 63.74Improved
InMaP on large encoderCLIP ViT-L/14@336 · ImageNetaccuracy (%): 77.02 -> 80.21Improved
add Sinkhorn-refined pseudo labels (ablation)InMaP^0.5 · ImageNet ResNet-50accuracy (%): 60.95 -> 63.74Improved
Sinkhorn labels without proxy learning (ablation)CLIP baseline · ImageNet ResNet-50accuracy (%): 60.32 -> 62.53Improved
proxy learning without refined labels (ablation)CLIP baseline · ImageNet ResNet-50accuracy (%): 60.32 -> 60.83Improved
raise tau_I from 0.01 to 0.04 (ablation)InMaP · ImageNet ResNet-50accuracy (%): 62.91 -> 63.74Improved
raise tau_I from 0.04 to 0.05 (ablation)InMaP · ImageNet ResNet-50accuracy (%): 63.74 -> 63.29Hurt
make all pseudo labels one-hot (alpha=0) (ablation)InMaP · ImageNet ResNet-50accuracy (%): 63.74 -> 63.14Hurt
vary threshold alpha between 1 and 0.5 (ablation)InMaP · ImageNet ResNet-50accuracy (%): 63.49 -> 63.72No clear effect
gamma=1 (no label distribution balancing) versus gamma=0.1 (ablation)InMaP · ImageNet ResNet-50accuracy (%): 63.75 -> 60.95Hurt
single prompt versus ensemble (ablation)InMaP · ImageNet ResNet-50accuracy (%): 63.15 -> 63.74Mixed
single prompt versus ensemble (ablation)Baseline · ImageNet ResNet-50accuracy (%): 58.15 -> 60.32Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy